Understanding Users with Disabilities

Proceedings of HF 2002, Nov. 25-27, 2002, Melbourne, Australia

Understanding Users with Disabilities’ Online Experience via a Discursive Approach

Natilene Bowker and Keith Tuffin

School of Psychology, Massey University, Palmerston North, New Zealand
natilene.bowker@actrix.co.nz, K.Tuffin@massey.ac.nz

Keywords: Research approaches, discursive psychology, identity, disability.

Abstract

This paper presents another approach for researchers to access users’ understandings of their online experience. The merits of discursive research - a particular kind of qualitative data gathering philosophy and technique, are considered. An empirical example of discursive research is given, which analyses users with disabilities’ (UWD) interview data about their online experiences. Research findings demonstrate the benefits of engaging in text-based environments for UWD because of users’ ability to control the textual construction of their identity. Considerations are suggested for multi-media design of online environments, where audio and visual enhancements are encouraged. Analysis offers a way into users’ conceptualisation of their online interaction, allowing users’ ideas to influence the research sphere. Beyond a functional capacity, the online medium is embedded within social practices, which enable users to operate beyond the constraints of a visible category of identity.

1. Approaches for investigating users’ online experiences

The online medium offers a social space for accessing information about how users operate online and, hence, in the social world. Users’ experiences can be accessed in different ways. Traditionally, research has sought to gain knowledge about online behaviour via quantitative methods like experiments (c.f. Dubrovsky, Kiesler, & Sethna, 1991; Kielser & Sproull, 1992; Rice, 1992; Sproull & Kiesler, 1986; Sproull, & Kiesler, 1991) and online surveys (c.f. Bowker & Liu, 2001; Bowker, in submission). Such approaches are particularly useful for testing and confirming hypotheses about users’ behaviour. To fully access the broad array of factors integral to understanding users’ online experience, however, which may extend beyond the researcher’s perspective and insight, it is important to consider qualitative approaches to gathering knowledge about how users operate online. Investigations conducted by Bowker (2001) demonstrate how the strengths of using qualitative approaches, inclusive of a thematic analysis of in-depth interviews and an ethnography, in conjunction with a quantitative study (an online survey) moved the research focus towards participants’ conceptualisation of identity exploration within an online community. While survey research affirmed the researcher’s belief in the importance of users’ real life gender in determining the degree of identity exploration, qualitative practices for gathering data illustrated the relevance of cultural stereotypes and the need to maintain stability in identity as significant factors constraining identity exploration online. The research identified the benefits of multiple approaches as a means of integrating both the researchers and participants’ situated knowledge, rather than reducing understandings about users’ online experience to uni-dimensional frameworks. Although the current paper does not deploy multiple approaches, it does focus on a particular kind of qualitative research strategy, namely discourse analysis (DA), which researchers can consider alongside standard research techniques utilised in the human computer interaction field. This kind of approach offers another way into accessing users’ interpretations of online technology.

DA has been adopted by social psychologists as a practical technique underpinned by a philosophical framework for understanding users’ conceptual constructions of their experiences. Within the field of psychology, DA has emerged alongside discursive psychology, which has developed from the traditions of semiotics, ethnomethodology, and linguistic philosophy (Potter & Wetherell, 1987; Edwards & Potter, 1992). Instead of privileging a measurable and observable representation of reality, as is the case with quantitative research methodologies, DA privileges language, because language or discourse is the basis of how we create meaning. Discursive psychology is grounded in the philosophy that knowledge changes according to the social and historical context. For instance, in mediaeval times people who did not follow the norm or were not happy with themselves were thought to be possessed by demons; in the 16th century states of melancholy were observed; in the early 20th century there were nervous breakdowns; in the later part of the 20th century identity crises occurred (Gergen, 1997). How do we know whether all these explanations refer to the same phenomena or something entirely different? How we interpret behaviour is determined by accepted cultural ideologies, constructed through the social practice of language, or discourse available within a particular context. As a consequence of language, which comprises the data for discursive researchers, being historically and contextually situated, DA focuses on people’s text as being representative of a particular social context, without the need to look at the representative nature of sample size, or ensure independence of data for the validity and reliability of outcomes.

Within a discursive philosophy language is primary. More than merely describing reality, language offers a means of accessing how people are choosing to conceptualise the social world, inclusive of the online medium. Language is seen to take an active and constitutive role in forming our experiences and understandings, including the experiences of online users. As a consequence of language being an integral part of any social context, discourse analysts do not assess validity on the basis of studying behaviour independently of the social context in which it is constructed (Wood & Kroger, 2000). Discourse analysts examine behaviour in the case of participants’ linguistic constructions, in connection with a particular social context because the social context is a valid component of the social phenomena under investigation. Validity in DA is evaluated according to how well grounded empirical evidence (in the form of textual utterances) occurs in support of the researcher’s claims.

Similarly, reliability in DA is not measured via the repeatability of measurable operations or conditions across time, samples, measures, and raters. Repetition is based on the repeatability of concepts or meanings. Reliability, in addition to validity, occurs through the rigorous and systematic analysis of grouping extracts of textual data into meaningful categories. These categories, limited by what participants say whether that be in the form of transcribed interview data or online textual data, function to demonstrate the conceptual practices or beliefs participants use to understand their experience, position, or situation. Documenting the process of analysis and extracts used provides readers an opportunity to independently evaluate the researcher’s analytic conclusions. This also allows readers to ensure that interpretations are consistent with the available data (Wood & Kroger, 2000). Offering an interpretation of the extracts allows the researcher to demonstrate how their interpretation, inclusive of an overall argument and its sub-claims, are grounded in the data (Wood & Kroger, 2000).

Discursive work has the advantage of gaining access to users’ online conceptual practices by providing a fine-grained analysis of their computer-mediated communication. This is achieved by looking at how language functions within a particular social context, and to what social ends it achieves (Edwards & Potter, 1992). Typically, this involves identifying the discursive patterns, which may appear in the form of repertoires and metaphors across participants’ text. Analysis involves considering the kinds of metaphors and repertoires users draw on to construct their experience as a means of accessing users’ conceptualisation of that experience.

Several advantages are offered for understanding users’ online experiences through a discursive approach. Drawing on the insights of Pauly (1991), Jones (1999a) suggests that in order to understand how people go about producing cultural forms of meaning-making, it is useful to conceptualise computer-mediated communication as a social practice. DA offers a useful insight in this regard because it focuses on how language functions as a social practice. Secondly, DA offers a means of accessing online experience in relation to everyday social practices. A number of researchers have highlighted the importance of studying the online medium in association with everyday life, rather than conceptualising it as an independent social space (Benedikt, 1991; Sterne, 1999; Jones, 1999a, 1999b; Kendall, 1999). Deploying a binary interpretation of on- and offline interaction offers a narrow method of understanding (Jones, 1999a). The notion that you can somehow cut off a part of your life and experience it independently of what has happened before or in other contexts is unrealistic according to Horn (1998). Discursive research, however, situates analysis within the context of everyday social practices (i.e. communication and language construction), and as such is free to consider users’ online experience in the context of everyday life.

Along with other techniques for gaining access to users’ interpretations of their online experience, discursive research may be well suited to the needs of online researchers who are interested in understanding the social and conceptual aspects of users’ engagement, within the context of everyday life. The following section offers an example of how a discursive approach can be deployed to access users’ conceptual constructions surrounding online identity.

2. Free from a disabled identity: How a socially marginalised user group conceptualises their online experience

This section provides an example of discursive research in practice by offering empirical data on how users conceptualise their online experience. The user group chosen for this research comprises members of a minority online user group. Data focuses on the ways users with disabilities (UWD) conceptualise their identity within a computer-mediated environment where visual cues are absent. Cromby and Standon (1999) propose that because the computer-mediated environment encourages textual self-presentation, UWD can interact in a medium where physical disability may be masked, bringing forth-greater flexibility in identity construction. This may offer possibilities for social and personal development.

2.1. Introduction

Turning to the background literature, studies of online communities have observed that opportunities for identity construction are made possible through textual communication. Reid’s (1991) pioneering research within the online community of Internet Relay Chat found the textual environment allowed users to operate outside traditional boundaries of race, gender, and age. In another study, Turkle (1995) observed that text-based role-playing games allowed users to explore various aspects of their identity and engage a wider range of roles. These studies demonstrate how online environments enable users to consider new ways of thinking about their identity as a textual construction, which is flexible.

Flexibility in identity construction may also lead to opportunities for UWD to position themselves without impairment affecting interaction. Hyde and Todd’s (1996) survey showed UWD felt the online medium offered them a social space where they could express their ideas without being judged on impairment. Similarly, interviews with UWD using computer technology in employment demonstrated the power of electronic communication in presenting a competent self (Roulstone, 1998). These examples highlight how technology allows UWD to be judged on their merits without prejudice.

2.2. Method

This research targeted people who identified as having a physical disability (inclusive of chronic illness) and, or a sensory disability. In clarifying what is meant by the term ‘disability’, the World Health Organization (2001: p.1) defines it as “an umbrella term for impairments, activity limitations or participation restrictions.” This study was conducted in New Zealand where 21 people with physical and sensory disabilities volunteered to be interviewed online via email, or another online communication program of their choice. Physical disabilities included cerebral palsy encompassing varying degrees of severity affecting speech, mobility, and co-ordination; brain injury; ataxia, a neurological disability resulting in lack of balance and co-ordination; shy-drager syndrome, a rare neurological disability causing degeneration in both the central and autonomic nervous systems; motor neurone disease; fibromyalgia, a chronic condition causing generalised muscular pain and fatigue; and post-polio syndrome. Sensory disabilities covered Deafness, visual impairment, and blindness. Participants ranged in age from 15 to 59 years and came from different walks of life with varying experiences in the workforce. Each participant used the online medium, inclusive of Internet and email, daily or at least several times a week.

2.3. Analysis

Interview data produced a total of 214 pages for analysis. Preliminary coding involved reading extracts multiple times to look for patterns in the data (Tuffin & Howard, 2001). Extracts, which talked about specific issues in a similar way, were grouped together. Data, which did not seem to fit into any categories, was left aside. Attempts were made to reintegrate these discarded items by expanding the initial categories for inclusivity. However, inevitably there were extracts, which did not fit into any categories except their own and, hence, were removed from the analysis. Revising and expanding initial categories allowed flexibility to consider alternative interpretations, which were later discarded due to lack of empirical evidence as supported by the data. Analysis took approximately 100 hours to complete, inclusive of preliminary coding. Qualitative data analysis software can be deployed to automate aspects of the analytic process. Researchers can search for all instances in the data where a specific word or phrase appears. These instances and their surrounding text can then be categorised according to appropriately defined concepts pertinent to the particular study. Qualitative software also provides a linear structure for storing all extracts, ensuring easy access to categories as well as individual extracts.

Amongst categories generated at a broad level was the idea of disclosing disability online. This was talked about as a matter of choice and became defined as a choice to disclose theme or repertoire (c.f. Bowker & Tuffin, 2002). An example of this repertoire follows. It is pertinent to note here that pseudonyms have been used to protect participants’ identities. As data has been copied verbatim from the text typed by participants online, any spelling and grammatical errors as well as ellipses have been retained in the extracts. In this extract Sarah is responding to a question about whether having a disability makes any difference online.

No it doesn't make a difference, but I don't have to bring it up if I don't want to! where as most of the time meeting someone in person the disability is spoken about at some stage. Online you can choose not to even bring it up. (Sarah)

Embedded within this repertoire is the idea that the presence or absence of disability is constructed as a feature controlled within the individual. Disclosing disability is cast as purely a matter of choice. Within the choice to disclose repertoire, three key resources were identified: relevance, anonymity, and normality. These resources appeared throughout participants’ data many times. Illustrative examples are given only.

2.4. Relevance resource

The relevance resource is built around the idea that disability becomes appropriate to disclose only when it is specific to the context of conversation. The following extract illustrates this.

I think it's only relevent when you need to establish your credibiity to talk about a subject relating to disability. For example, people don't need to know that I have a disability if I am contributing to an analysis of the last All Black test, fine wines and food, movies, raising teenagers etc because it has no relevence. What might be more relevent is my gender, age, race, socio-economic status, parental status, academic background, where I live, who I know etc etc etc. (Shaun)

Shaun mentions disability only when it is useful in justifying his knowledge about the subject. In all other cases it is irrelevant. In fact, other personal characteristics transcend disability.

2.5. Anonymity resource

Turning to the second resource used in the choice to disclose repertoire is anonymity. This resource is based on the idea that non-disclosure continues until one chooses to make personal information available. In the following extract, Bridgette highlights the positive aspects of being anonymous online.

Quite often they dont know I have a disability - so there is an anonomous element available to me, that is good! No need to explain or be at the brunt of poeples asssummptions ... in some cases I do reveal my disability, describing myself, and people are quite accepting, probably cos they cannot see me and thus dont get the full impact of my disabiliity. (Bridgette)

Bridgette defines anonymity as an ability to refrain from revealing a salient aspect of her physical identity. This is justified because prejudice is experienced when disability is exposed. Being anonymous removes Bridgette from having to deal with explanations and assumptions about her disability. Instead, assumptions of normality dominate. Unsettling these assumptions actually demands an act of identity construction. When Bridgette does reveal her disability, prejudice is softened due to lack of visual cues. Hence, even when anonymity is removed online, experience of disability may change. Bridgette claims people are more accepting of her disability because visual cues are unavailable.

2.6. Normality resource

The third resource identified in the choice to disclose repertoire is normality. The normality resource draws on non-disclosure as a tool, which allows UWD to engage in able-bodied forms of interaction. The following extract highlights the normality resource. Sally-Ann is responding to a question about whether there is a difference between online and face-to-face interactions.

Yes, I have found this to be so. I am able to regain what others would consider normal control. My computer allows me to do ordinary things without my disability getting in the way. When I think it is necessary or important for some reason I will let them know. I value being able to exercise that right and choice as it is not one I can exercise in my daily life. Being online enables me to enjoy the world as others do and it also means for the most part I can take a break mentally from disability. A luxury I reckon. (Sally-Ann)

The phrase “normal control” refers to Sally-Ann’s ability to gain control over disability disclosure within routine interactions. Disability is transformed into a characteristic of choice. Power over how it is presented is located within the individual. Having the ability to decide when disability is relevant to bring into discussion is talked about as a participatory right, which has typically been inaccessible to UWD. Having normal control over disclosure gives Sally-Ann the ability to enjoy the world on a par with the able-bodied. This experience is valued.

Being online also gives Sally-Ann the opportunity to take a ‘mental break’ from disability. Locating disability within a mental dimension means Sally-Ann can step outside its physical constraints. Constructing her online experience as taking a ‘mental’ break suggests she is not denying the material existence of disability. Rather, from a conceptual perspective only, disability does not need to be accessed, so she is afforded a conceptual holiday. In the final sentence, the term “luxury” denotes that having access to a space where disability can be removed mentally is rare and is also highly valued.

To summarise, the textual medium removes the dominance of a visual interactive mode, which has traditionally governed the way people interpret and evaluate others. This affords UWD agency over important aspects of their identity. Presenting as disabled becomes a matter of choice controlled within the individual. Controlling when and how disability is engaged, coupled with the ability to flexibly construct an identity, are opportunities, which have often been inaccessible to UWD. The value UWD place on being able to control when, where, and how their disabled identity is constructed in online interaction presents some interesting outcomes and considerations.

2.7. Outcomes and considerations

Firstly, with respect to the particular user group investigated, the discursive findings suggest some considerations for online researchers producing policy guidelines for designing online applications. While there has been a trend towards enriching the bandwidth of computer-mediated communication via multi-media technology, which integrates audio and visual enhancements for online interaction, text-based environments may be more attractive to UWD because of their flexibility in identity construction. Introducing opportunities for accessing audio and visual stimuli about other users in a communicative context may serve to inhibit participation from those whose physical and sensory abilities differ from the majority.

Further, the social and conceptual frameworks deployed by users who contributed to this research may function to position the online medium as a social space available for engaging in activities where identity occupies a flexible and constructive dimension. In this regard, users may value the nature of online environments because they are distinct from face-to-face contexts where audio and visual stimuli construct and, hence, also, constrain reality. UWD may particularly value software as well as hardware, which does not expose differences in co-ordination (such as lack of manual dexterity and slow typing speed) and communication abilities (inclusive of Deaf users, who may be proficient in Sign language, but not the principal language of online communication, namely English). These considerations can be extrapolated to other online user groups whose (visually apparent) social identity may create negative and disadvantageous outcomes.

In relation to UWD’s online preferences, potential considerations may extend to online accessibility issues. If UWD value the ability to refrain from disclosing a disabled identity online, a design practice involving separately accessible web products or web sites purely catering to the needs of UWD may be less attractive to this particular user group. Rather than developing alternative systems, it may be better to integrate accessibility features for UWD into mainstream environments.

The discursive interpretations highlight how the online medium, beyond offering an accessible environment for functioning within, may provide an accessible social space where users can leave behind their social identities, and, thus, the prejudice reactions evoked by such visually apparent categories. Discursive findings also indicate that while prejudice may not be encountered online, once relevant details are made available, discriminatory judgements can be accessed and realised.

3. Conclusion

A discursive approach to understanding how users operate online may present considerations for researchers (and in some cases designers) wanting to access the social and conceptual aspects of computer-human interaction. As outlined in the research example, a discursive analysis of users’ computer-mediated communication produced another way into the conceptual frameworks users deployed to operate within online environments. Data analysis gave access to the social meaning surrounding users’ online experience. More than providing a functional capacity for operating, the online medium is embedded within social practices of relevance, anonymity, and normality, which enable users to operate beyond the constraints and prejudice associated with a visibly apparent category of identity.

4. References

· Benedikt, M. (1991) Cyberspace: Some Proposals, in Benedikt, M. (ed.) Cyberspace: First Steps, 119-224, (MIT Press: Cambridge).

· Bowker, N.I. (2001) Understanding Online Communities through Multiple Methodologies Combined under a Postmodern Research Endeavour, Forum Qualitative Sozialforschung / Forum: Qualitative Social Research [Online], 2(1). Available: http://qualitative-research.net/fqs/fqs-eng.htm

· Bowker, N.I. (in submission) Gender-Stereotyped Beliefs Correspond to the Behavioural Preferences of Experienced Chatroom Users, Personality and Social Psychology Bulletin.

· Bowker, N.I.; Liu, J. (2001) Are Women Occupying Positions of Power On-line?: Demographics of Chat Room Operators, CyberPsychology and Behavior: The Impact of the Internet, Multimedia and Virtual Reality on Behavior and Society, 4(5), 631-644.

· Bowker, N.I; Tuffin, K. (2002) Disability Discourses for Online Identities, Disability & Society, 17(3), 327-344.

· Cromby, J.; Standon, P. (1999) Cyborgs and Stigma: Technology, Disability, Subjectivity, in Gordo-Lopez, A.J. and Parker, I. (eds.) Cyberpsychology, pp. 95-112, (Routledge: New York).

· Dubrovsky, V.; Kiesler, S. and Sethna, B. (1991) The Equalization Phenomenon: Status Effects in Computer-Mediated and Face-to-Face Decision-Making Groups, Human Computer Interaction, 6, 119-146.

· Edwards, D.; Potter, J. (1992) Discursive Psychology (Sage: London).

· Gergen, K.F. (1997) Social Psychology as Social Construction: The Emerging Vision, in McGarty, C. and Haslam, S.A. (eds.) The Message of Social Psychology: Perspectives on Mind in Society, 113-128, (Blackwell: Oxford).

· Horn, S. (1998) Cyberville: Clicks, Culture and the Creation of an Online Town, (Warner Books: New York).

· Hyde, D.; Todd, R. (1996) An Overview of Issues Surrounding Use of the Internet by People with Disabilities, Proceedings of the 18th World Congress of Rehabilitation International, (Auckland, New Zealand).

· Jones, S. (1999a) Studying the Net: Intricacies and Issues, in Jones, S. (ed.) Doing Internet research: Critical Issues and Methods for Examining the Net, 1-27, (Sage: Thousand Oaks, CA.).

· Jones, S. (ed.) (1999b) Doing Internet Research: Critical Issues and Methods for Examining the Net, (Sage: Thousand Oaks, CA.).

· Kendall, L. (1999) Recontextualising “Cyberspace”: Methodological Considerations for On-line Research, in Jones, S. (ed.) Doing Internet research: Critical Issues and Methods for Examining the Net, 57-74, (Sage: Thousand Oaks, CA.).

· Kielser, S.; Sproull, L. (1992) Group Decision Making and Communication Technology, Organizational Behavior and Human Decision Processes, 52, 96-123.

· Pauly, J. J. (1991) A Beginner’s Guide to Qualitative Research, Journalism Monographs (serial no. 125).

· Potter, J.; Wetherell, M. (1987) Discourse and Social Psychology: Beyond Attitudes and Behaviour, (Sage: London).

· Reid, E. (1996) Informed Consent in the Study of On-line Communities: A Reflection on the Effects of Computer-Mediated Social Research, The Information Society, 12(2), 169-174.

· Rice, R. (1992) Contexts of Research on Organizational Computer-Mediated Communication, in Lea, M. (ed.) Contexts of Computer-Mediated Communication, 113-145, (Harvester-Wheatsheaf: London).

· Roulstone, A. (1998) Researching a Disabling Society: The Case of Employment and New Technology, in Shakespeare, T. (ed) The Disability Reader: Social Sciences Perspectives, 110-128, (Cassell: Great Britain).

· Sproull, L.; Kiesler, S. (1986) Reducing Social Context Cues: Electronic Mail in Organisational Comunication, Management Science, 32, 1492-1512.

· Sproull, L.; Kiesler, S. (1991) Connections: New Ways of Working in the Networked Environment, (MIT Press: Cambridge, MA).

· Sterne, J. (1999). Thinking the Internet: Cultural Studies versus the Millenium, in Jones, S. (ed.) Doing Internet research: Critical Issues and Methods for Examining the Net, 257-287, (Sage: Thousand Oaks, CA.).

· Tuffin, K; Howard, C. (2001) Demystifying Discourse Analysis: Theory, Method and Practice, in McHoul, A. and Rapley, M. (eds.) How to Analyse Talk in Institutional Settings: A Casebook of Methods, 196-205, (Continuum International: London).

· Turkle, S. (1995) Life on the Screen: Identity in the Age of the Internet (Simon & Schuster, New York).

· Wood, L.A.; Kroger, R.O. (2000) Doing Discourse Analysis: Methods for Studying Action in Talk and Text, (Sage: Thousand Oaks, CA).

· World Health Organization. (2001) International Classification of Functioning, Disability and Health (2nd ed.), (Classification, Assessment, Surveys and Terminology Team, Author, Geneva, Switzerland).